Version Of Particle Swarm Optimization Research Articles

Fuzzy Self-Tuning PSO: A settings-free algorithm for global optimization

Abstract Among the existing global optimization algorithms, Particle Swarm Optimization (PSO) is one of the most effective methods for non-linear and complex high-dimensional problems. Since PSO performance strongly depends on the choice of its settings (i.e., inertia, cognitive and social factors, minimum and maximum velocity), Fuzzy Logic (FL) was previously exploited to select these values. So far, FL-based implementations of PSO aimed at the calculation of a unique settings for the whole swarm. In this work we propose a novel self-tuning algorithm—called Fuzzy Self-Tuning PSO (FST-PSO)—which exploits FL to calculate the inertia, cognitive and social factor, minimum and maximum velocity independently for each particle, thus realizing a complete settings-free version of PSO. The novelty and strength of FST-PSO lie in the fact that it does not require any expertise in PSO functioning, since the behavior of every particle is automatically and dynamically adjusted during the optimization. We compare the performance of FST-PSO with standard PSO, Proactive Particles in Swarm Optimization, Artificial Bee Colony, Covariance Matrix Adaptation Evolution Strategy, Differential Evolution and Genetic Algorithms. We empirically show that FST-PSO can basically outperform all tested algorithms with respect to the convergence speed and is competitive concerning the best solutions found, noticeably with a reduced computational effort.

Formulation and parameter selection of multi-objective deterministic particle swarm for simulation-based optimization

Global derivative-free deterministic algorithms are particularly suitable for simulation-based optimization, where often the existence of multiple local optima cannot be excluded a priori, the derivatives of the objective functions are not available, and the evaluation of the objectives is computationally expensive, thus a statistical analysis of the optimization outcomes is not practicable. Among these algorithms, particle swarm optimization (PSO) is advantageous for the ease of implementation and the capability of providing good approximate solutions to the optimization problem at a reasonable computational cost. PSO has been introduced for single-objective problems and several extension to multi-objective optimization are available in the literature. The objective of the present work is the systematic assessment and selection of the most promising formulation and setup parameters of multi-objective deterministic particle swarm optimization (MODPSO) for simulation-based problems. A comparative study of six formulations (varying the definition of cognitive and social attractors) and three setting parameters (number of particles, initialization method, and coefficient set) is performed using 66 analytical test problems. The number of objective functions range from two to three and the number of variables from two to eight, as often encountered in simulation-based engineering problems. The desired Pareto fronts are convex, concave, continuous, and discontinuous. A full-factorial combination of formulations and parameters is investigated, leading to more than 60,000 optimization runs, and assessed by three performance metrics. The most promising MODPSO formulation/parameter is identified and applied to the hull-form optimization of a high-speed catamaran in realistic ocean conditions. Its performance is finally compared with four stochastic algorithms, namely three versions of multi-objective PSO and the genetic algorithm NSGA-II.

An accelerated PSO for query expansion in web information retrieval: application to medical dataset

Swarm intelligence algorithms are now among the most widely used soft computing techniques for optimization and computational intelligence. One recent swarm intelligence algorithm that has begun to receive more attention is Accelerated Particle Swarm Optimization (APSO). It is an enhanced version of PSO with global optimization capability, sufficient simplicity and high flexibility. In this paper, we propose the application of the APSO technique to efficiently solve the problem of Query Expansion (QE) in Web Information Retrieval (IR). Unlike prior studies, we introduce a new modelling of QE that aims to find the suitable expanded query from among a set of expanded query candidates. Nevertheless, due to the large number of potential expanded query candidates, it is extremely complex to produce the best one through conventional hard computing methods. Therefore, we propose to consider the problem of QE as a combinatorial optimization problem and address it with APSO. We thoroughly evaluate the proposed APSO for QE using MEDLINE, the world Web’s largest medical library. We first conduct a preliminary experiment to tune the APSO parameters. Then, we compare the results to a recent swarm intelligence algorithm called Firefly Algorithm (FA). We also compare the results with three recently published methods for QE that involved Particle Swarm Optimization (PSO), Genetic Algorithm (GA) and Bat Algorithm (BA). The experimental analysis demonstrates that the proposed APSO for QE is very competitive and yields substantial improvement over the other methods in terms of retrieval effectiveness and computational complexity.

Multilevel thresholding using grey wolf optimizer for image segmentation

Abstract Multilevel thresholding is one of the most important areas in the field of image segmentation. However, the computational complexity of multilevel thresholding increases exponentially with the increasing number of thresholds. To overcome this drawback, a new approach of multilevel thresholding based on Grey Wolf Optimizer (GWO) is proposed in this paper. GWO is inspired from the social and hunting behaviour of the grey wolves. This metaheuristic algorithm is applied to multilevel thresholding problem using Kapur's entropy and Otsu's between class variance functions. The proposed method is tested on a set of standard test images. The performances of the proposed method are then compared with improved versions of PSO (Particle Swarm Optimization) and BFO (Bacterial Foraging Optimization) based multilevel thresholding methods. The quality of the segmented images is computed using Mean Structural SIMilarity (MSSIM) index. Experimental results suggest that the proposed method is more stable and yields solutions of higher quality than PSO and BFO based methods. Moreover, the proposed method is found to be faster than BFO but slower than the PSO based method.

All-dimension neighborhood based particle swarm optimization with randomly selected neighbors

Particle swarm optimization (PSO) is widely used for solving various optimization problems, since it has few parameters and is easy to implement. However, canonical PSO generally suffers from premature convergence because it usually loses diversity too rapidly during the evolutionary process. To improve the performance of PSO on complex problems, an all-dimension-neighborhood-based PSO with randomly selected neighbors learning strategy (ADN-RSN-PSO) is proposed in this study. The randomly selected neighbors (RSN) learning strategy is adopted in the early stage of PSO to enhance the swarm diversity, while the all-dimension neighborhood (ADN) strategy is utilized in the later stage to accelerate the convergence rate. The ADN strategy enhances the local search capability around the global-best solution in a dimension by dimension manner, and the search distance is adapted by shrinking and random-expansion operators. Experimental results show that ADN-PSO can improve the exploitation capability of the global version of PSO. To test the performance of the proposed ADN-RSN-PSO, comparison tests on the CEC2013 test suite are carried out. The comparison results reveal that ADN-RSN-PSO outperforms other peer PSO variants. In the end, the proposed ADN-RSN-PSO is applied to the radar system design problem to demonstrate its potential in real-life applications.

Marker-based human pose tracking using adaptive annealed particle swarm optimization with search space partitioning

Pose estimation and tracking of an articulated structure based on data from multiple cameras has seen numerous applications in recent years. In this paper, a marker-based human pose tracking algorithm from multi view video sequences is proposed. The purpose of the proposed algorithm is to present a low cost motion capture system that can be used as an alternative to high cost available commercial human motion capture systems. The problem is defined as the optimization of 45 parameters which define body pose model and is solved using a modified version of particle swarm optimization (PSO) algorithm. The objective of this optimization is to maximize a fitness function which formulates how much the body model matches with 2D marker coordinates in video frames. A sampling covariance matrix is used in the first part of the velocity equation of PSO and is annealed with iterations. The sampling covariance matrix is computed adaptively, based on variance of parameters in the swarm. One of the concerns in this algorithm is the high number of parameters to define the model of body pose. To tackle this problem, we partition the optimization state space into six stages that exploit the hierarchical structure of the skeletal model. The first stage optimizes the six parameters that define the global orientation and position of the body. Other stages relate to optimization of right and left hand, right and left leg and head orientation. In the proposed partitioning method previously optimized parameters are allowed some variation in each step that is called soft partitioning. Experimental results on Pose Estimation and Action Recognition (PEAR) database indicate that the proposed algorithm achieves lower estimation error in tracking human motion compared with Annealed Particle Filter (APF) and Parametric Annealing (PA) methods.

A hybrid Particle Swarm Optimization – Variable Neighborhood Search algorithm for Constrained Shortest Path problems

In this paper, a well known NP-hard problem, the Constrained Shortest Path problem, is studied. As efficient metaheuristic approaches are required for its solution, a new hybridized version of Particle Swarm Optimization algorithm with Variable Neighborhood Search is proposed for solving this significant combinatorial optimization problem. Particle Swarm Optimization (PSO) is a population-based swarm intelligence algorithm that simulates the social behavior of social organisms by using the physical movements of the particles in the swarm. A Variable Neighborhood Search (VNS) algorithm is applied in order to optimize the particles’ position. In the proposed algorithm, the Particle Swarm Optimization with combined Local and Global Expanding Neighborhood Topology (PSOLGENT), a different equation for the velocities of particles is given and a novel expanding neighborhood topology is used. Another issue in the application of the VNS algorithm in the Constrained Shortest Path problem is which local search algorithms are suitable from this problem. In this paper, a number of continuous local search algorithms are used. The algorithm is tested in a number of modified instances from the TSPLIB and comparisons with classic versions of PSO and with other versions of the proposed method are performed. Also, the results of the algorithm are compared with the results of a number of metaheuristic and evolutionary algorithms. The results obtained are very satisfactory and strengthen the efficiency of the algorithm.

A Novel Distributed Quantum-Behaved Particle Swarm Optimization

Quantum-behaved particle swarm optimization (QPSO) is an improved version of particle swarm optimization (PSO) and has shown superior performance on many optimization problems. But for now, it may not always satisfy the situations. Nowadays, problems become larger and more complex, and most serial optimization algorithms cannot deal with the problem or need plenty of computing cost. Fortunately, as an effective model in dealing with problems with big data which need huge computation, MapReduce has been widely used in many areas. In this paper, we implement QPSO on MapReduce model and propose MapReduce quantum-behaved particle swarm optimization (MRQPSO) which achieves parallel and distributed QPSO. Comparisons are made between MRQPSO and QPSO on some test problems and nonlinear equation systems. The results show that MRQPSO could complete computing task with less time. Meanwhile, from the view of optimization performance, MRQPSO outperforms QPSO in many cases.

A Configurable Routing Protocol for Improving Lifetime and Coverage Area in Wireless Sensor Networks

The particularities of Wireless Sensor Networks require specially designed protocols. Nodes in these networks often possess limited access to energy (usually supplied by batteries), which imposes energy constraints. Additionally, WSNs are commonly deployed in monitoring applications, which may intend to cover large areas. Several techniques have been proposed to improve energy-balance, coverage area or both at the same time. In this paper, an alternative solution is presented. It consists of three main components: Fuzzy C-Means for network clustering, a cluster head rotation mechanism and a sleep scheduling algorithm based on a modified version of Particle Swarm Optimization. Results show that this solution is able to provide a configurable routing protocol that offers reduced energy consumption, while keeping highcoverage area.

Multi-Objective Distribution Network Reconfiguration Based on Pareto Front Ranking

Electrical distribution system reconfiguration is frequently addressed as a multi-objective problem, typically taking into account the system losses together with other objectives, among which reliability indicators are widely used. In the multi-objective context, Pareto front analysis enables the operator handling conflicting and even non-commensurable objectives without needing the use of additional hypotheses or weights. This paper provides advances on the application of Pareto front analysis to multi-objective distribution network reconfiguration. Starting from previous results in which genetic algorithms were effectively adopted to find the best-known Pareto front, a version of the multi-objective binary particle swarm optimization (MOBPSO) customized for distribution network reconfiguration has been developed by exploiting the internal ranking of the solutions (based on a multi-criteria decision making method in the selection of the local best) and the network topology. Furthermore, the Pareto front mismatch metric (already used by the authors to compare different methods for small networks for which the complete Pareto front can be calculated) has been generalized to be used with large systems for which only the best-known Pareto front is found. Applications to a test network and to a real urban distribution network are discussed, showing the consistent superiority of the customized MOBPSO version with respect to the application of genetic algorithms and of a more classical version of the particle swarm optimization method.

An Intelligent Novel Tripartite - (PSO-GA-SA) Optimization Strategy

This paper presents a tripartite version of particle swarm optimisation, genetic algorithm, and simulated annealing (PSO-GA-SA) optimisation strategy addressing some predominant issues such as the problem of the potential solution being trapped in a local minima solution space, the untimely convergence and the slow rate of arriving at optimal solutions. This strategy is designed with an intelligence beneficiary trade-off between exploration and exploitation of the full potential of all the capabilities of PSO, GA, and SA functioning simultaneously. The design algorithm further incorporates a variable velocity component that introduces random intelligence. There are substantial performance improvements when the novel robust design is first validated with three test functions for the initial case studies. To demonstrate the capabilities to handle complexities and establish scalability in the implementation of the proposed approach, the optimisation strategy is further applied to a high-integrity protection system (HIPS) which is a real-life safety system design optimisation problem with increased number of input variables, constraints, and limitations on the available resources. The novel design performs better than their individual methods using the number of fitness evaluations, as the performance metrics, whilst operating with both a reduced number of generations and initial number of starting potential solutions.

Spider monkey optimization algorithm for constrained optimization problems

In this paper, a modified version of spider monkey optimization (SMO) algorithm for solving constrained optimization problems has been proposed. To the best of author's knowledge, this is the first attempt to develop a version of SMO which can solve constrained continuous optimization problems by using the Deb's technique for handling constraints. The proposed algorithm is named constrained spider monkey optimization (CSMO) algorithm. The performance of CSMO is investigated on the well-defined constrained optimization problems of CEC2006 and CEC2010 benchmark sets. The results of the proposed algorithm are compared with constrained versions of particle swarm optimization, artificial bee colony and differential evolution. Outcome of the experiment and the discussion of results demonstrate that CSMO handles the global optimization task very well for constrained optimization problems and shows better performance in comparison with compared algorithms. Such an outcome will be an encouragement for the research community to further explore the potential of SMO in solving benchmarks as well as real-world problems, which are often constrained in nature.

Synthesis of Unequally Spaced Linear Micro Strip Rectangular Patch Antenna Array Using Improved Local Search Particle Swarm Optimization

Antenna array systems with low side lobe levels are essential for today wireless communication systems. This paper presents the synthesis of unequally spaced linear rectangular micro strip antenna array with minimum side lobe levels using the novel evolutionary algorithm known as improved local search particle swarm optimization (ILSPSO). ILSPSO is a modified version of particle swarm optimization (PSO), in which Gaussian distribution is used to enhance the local search of the PSO. In this paper, ILPSO is applied to optimize the positions of the micro strip antenna elements to suppress the peak side lobe level (PSLL) along with PSO and differential evolution (DE) algorithms. The steps involved in problem formulation along with design examples illustrating the performance of the ILPSO in minimizing the side lobe levels are demonstrated. A 20 and 32 element linear micro strip rectangular patch antenna (MSRPA) element are considered to show the effectiveness of the proposed method. The optimized micro strip antenna array is simulated using high frequency structure simulator (HFSS). The synthesis results demonstrate that the ILSPSO outperforms PSO and DE in terms of producing lower PSLL and convergence rate. The flexibility and ease of implementation of the ILSPSO algorithm is obvious from this paper, showing the algorithms usefulness in other array synthesis problems.

Interleaving of particle swarm optimization and differential evolution algorithm for global optimization

Stochastic optimization algorithms have potential to solve optimization problems in various fields of engineering and science. However, increasing non-linearity, non convexity, multi-modality, discontinuity, and even dynamics make the problems more complex and intractable. Classical optimization techniques are not able to determine global solution by analyzing rough non-linear surfaces. Heuristic algorithms have been used for determining global solution for this type of problems. However, heuristic algorithm is knowledge dependent, so finding a unique heuristic optimization algorithm for obtaining optimum solutions for all problems is not feasible. Hybridization is an integrated framework where merits of algorithms are utilized to improve performance of the optimizers. Particle Swarm Optimization (PSO) and Differential Evolution (DE) algorithm are two heuristic algorithms despite certain shortcomings have been applied to solve global optimization problems. In this paper, we propose an integrated framework where improved version of PSO and DE (IPSODE) is executed in interleaved fashion for balancing exploration and exploitation dilemma in the evolution process. In IPSODE algorithm, generation starts with improved PSO and switch to either improved DE or continue with improved PSO based on the fitness value. The algorithm is experimented on 20 benchmark/test functions which are uni modal, multi-modal, shifted, and rotated with 10, 20, and 30 different dimensions. The performance of the proposed method is confirmed by comparing with basic PSO, basic DE, and three hybridization methods of PSO and DE based on evaluation criteria like solution quality, robustness, convergence speed, scalability, and statistical t-test.

Optimal Sizing of Standalone Wind/PV/Fuel Cell/Battery System Using Adaptive Particle Swarm Optimization

Hybrid energy systems play a key role in sustainable development. This paper presents a Techno-Economic evaluation of the problem of optimal sizing for a windsolar system, including a hybrid back-up which consists of a fuel cell and a battery bank. With the help of mathematical models of the wind turbines, solar panels, fuel cells and lead-acid batteries, this paper reports an intelligent methodology for optimal sizing using Adaptive particle swarm optimization. The energy management strategy is innovative but simple and performs efficient hybrid backup management as well as the overall system management. The problem has initially been solved with a conventional iterative procedure (unit increment algorithm) in order to serve as a reference, but later an adaptive version of particle swarm optimization (APSO) is used. Reliability and Per unit cost of a hybrid energy system has been considered as the key parameter for evaluating the techno-economical feasibility, where the number of each equipment and the volume of hydrogen has been considered as the variables for optimization.

An Optimized K-Harmonic Means Algorithm Combined with Modified Particle Swarm Optimization and Cuckoo Search Algorithm

Abstract Among the data clustering algorithms, k-means (KM) algorithm is one of the most popular clustering techniques due to its simplicity and efficiency. However, k-means is sensitive to initial centers and it has the local optima problem. K-harmonic-means (KHM) clustering algorithm solves the initialization problem of k-means algorithm, but it also has local optima problem. In this paper, we develop a new algorithm for solving this problem based on an improved version of particle swarm optimization (IPSO) algorithm and KHM clustering. In the proposed algorithm, IPSO is equipped with Cuckoo Search algorithm and two new concepts used in PSO in order to improve the efficiency, fast convergence and escape from local optima. IPSO updates positions of particles based on a combination of global worst, global best with personal worst and personal best to dynamically be used in each iteration of the IPSO. The experimental result on five real-world datasets and two artificial datasets confirms that this improved version is superior to k-harmonic means and regular PSO algorithm. The results of the simulation show that the new algorithm is able to create promising solutions with fast convergence, high accuracy and correctness while markedly improving the processing time.

An automated approach for optimal design of prestressed concrete slabs using PSOHS

In this paper a new automated approach is presented for optimal design of prestressed concrete slabs. To achieve this goal, the model of a slab is formed using SAP2000, and it is linked to a meta-heuristic code. This code utilizes the result of analyzed models in each iteration to provide new design parameters of the slab for the subsequent one. Canadian Standard Association requirements are met thoroughly to reach a safe code-based design. A recently enhanced version of Particle Swarm Optimization (PSO), so-called PSOHS, is employed, and it is shown that the latter is superior to the standard PSO. Results prove the efficiency of the PSOHS, indicating the possibility of its application in professional engineering. Furthermore, the comparison of the PSO and PSOHS shows that PSOHS is less parameter sensitive, and provides final designs with smaller cost functions.

Routing algorithm with QoS constraints for real time multicast communications using particle swarm optimization and genetic meta-optimizer

Purpose The current work shows an approach to solve the QoS multicast routing problem by using Particle Swarm Optimization (PSO). The problem of finding a route from a source node to multiple destination nodes (multicast) at a minimum cost is an NP-Complete problem (Steiner tree problem) and is even greater if Quality of Service -QoS- constraints are taken into account. Thus, approximation algorithms are necessary to solve this problem. This work presents a routing algorithm with two QoS constraints (delay and delay variation) for solving the routing problem based on a modified version of particle swarm optimization. Design/methodology/approach This work involved the following methodology: 1. Literature Review 2. Routing algorithm design 3. Implementation of the designed routing algorithm by java programming. 4. Simulations and results. Findings In this work we compared our routing algorithm against the exhaustive search approach. The results showed that our algorithm improves the execution times in about 40% with different topologies. Research limitations/implications The algorithm was tested in three different topologies with 30, 40 and 50 nodes with and a dense graph topology. Originality/value Our algorithm implements a novel technique for fine tuning the parameters of the implemented bio-inspired model (Particles Swarm Optimization) by using a Genetic Meta-Optimizer. We also present a simple and multi implementation approach by using an encoding system that fits multiple bio-inspired models.

Optimal sizing and design of CMOS analogue amplifier circuits using craziness‐based particle swarm optimization

SummaryThis paper presents the optimal designs of two analogue complementary metal–oxide–semiconductor (CMOS) amplifier circuits, namely differential amplifier with current mirror load and two‐stage operational amplifier. A modified Particle Swarm Optimization (PSO), called Craziness‐based Particle Swarm Optimization (CRPSO) technique is applied to minimize the total MOS area of the designed circuits. CRPSO is a highly modified version of conventional PSO, which adopts a number of random variables and has a better and faster exploration and exploitation capability in the multidimensional search space. Integration of craziness factor in the fundamental velocity term of PSO not only brings diversity in particles but also pledges convergence close to global best solution. The proposed CRPSO‐based circuit optimization technique is reassured to be free from the intrinsic disadvantages of premature convergence and stagnation, unlike Differential Evolution (DE), Harmony Search (HS), Artificial Bee Colony (ABC) and Particle Swarm Optimization (PSO). The simulation results achieved for the two analogue CMOS amplifier circuits establish the efficacy of the proposed CRPSO‐based approach over those of DE, HS, ABC and PSO in terms of convergence haste, design conditions and design goals. The optimally designed analogue CMOS amplifier circuits occupy the least MOS area and show the best performance parameters like gain and power dissipation, in compared with the other reported literature. Copyright © 2016 John Wiley & Sons, Ltd.

Co-Kriging Assisted PSO Algorithm and Its Application to Optimal Transposition Design of Power Transformer Windings for the Reduction of Circulating Current Loss

An optimal transposition of power transformer windings is proposed to minimize the circulating current loss by incorporating co-Kriging into a binary version of particle swarm optimization. The co-Kriging takes most samplings from less accurate but cheap data, while taking only some from accurate but expensive data to increase its fitting accuracy and to reduce the computing cost at the same time. The effectiveness of the proposed algorithm is demonstrated through an analytical example, and applied to an optimal transposition design of a power transformer to minimize the circulating current loss.